Tuesday, 5 June 2012

Reality Disproved

Eventually, in a narrative of the artificial intelligence singularity, computers become so advanced that desktop PCs are able to process with incredible speed beyond the collective brain-power of every man, woman, and child who ever lived.  And if this were the case, a proposal known as "simulation theory" suggests that the chances of being a product of one of these simulations is almost guaranteed.  So how did one Oxford University Professor call everything in reality into question?
The paper at the centre of the controversy is titled "Are You Living in a Computer Simulation?"  Nick Bostrom, award winning Swedish Philosopher and professor may have come dangerously close to disproving reality as we know it.  Circulated in 2003, the paper goes down the rabbit hole to suggest that we are not, as many have taken for granted, in a physical world governed by immutable laws that transcend the fabric of the universe dictated by forces unknown at its periphery. 
Instead, it puts forward the following three possible scenarios given today's already emerging technological trends.  First, it says it is quite possible that the human race will inevitably go extinct before it achieves a "posthuman" stage where artificial intelligence is able to improve upon itself in a Ray Kurzweil style AI singularity.  If that is not the case, it suggests the idea that some factor in the human race's future will prevent more than a handful of advanced computer simulations of its own past from being built.  And the final scenario it puts forward is the most shocking of all - that we are currently living in an environment generated entirely by a computer.  And that's not all.  According to the paper, even our very minds may be simply byproducts of the computer simulation.  It states that rather than a mere possibility, the likelihood of our being made from information is surprisingly high.
One of the interesting things about this paper is the hardware/software barrier of our current understanding of neuroscience doesn't enter into the equation.  It suggests that we may be wholly generated by the computer itself, consciousness and all.
There is a troubling subtext here that makes some fairly critical predictions about the future of humanity.  If a computer simulated reality of our world is feasible and within the scope of our future, then it's statistically likely that we are living in one right now.  If, however, we contend that we are not living in a computer simulated reality then it seems to follow that such a simulation may be impossible.  If it were possible to make a reality such as this using Earthly (or any other) technology, then the number of universes generated by such a method would quickly outnumber our understood single universe.  From there, it's a matter of how many computers an advanced society can create.
The concept of a simulated reality hinges on the likelihood of us ever creating one on Earth.  If we make one, then they are theoretically possible and we may very well live inside one already.  And that would mean our machine had created a simulation within a simulation like a Christopher Nolan film.
An additional thought exercise to explore while looking into this departs only slightly from the "ancestor simulation" idea.  If our universe had a starting point (whenever they switched the simulation on) we could look at this universe as something akin to a work of art that most certainly require universal laws to operate within the confines of the canvas.
The universe birthing this one, however, may have completely different laws.  Whereas A must come before B in this universe because this is how the code was written, in the universe creating this one B could very well come before A or simultaneously alongside A - even existing without A being part of the picture.  In this way, the parent universe could require no starting point and have no end to it, but could have origins self-evident to those living on it.
So rather than something coming from nothing, different laws from the parent universe could mean the previous universe operated outside our canvas-bound understanding of the cosmos and the nature of our painted reality.  It could be a "self-starter" or an "always was" that appears illogical to those of us operating under purely manufactured laws of physics in a simulated reality.  At this level of speculation, the line where reality ends and technology begins starts to blur - and could end up being more allegorical and magical than anything we currently understand in the harder sciences.
Of course this canvas analogy is difficult to imagine, and has more than a few problems.  It's one of those speculative ventures that stacks the deck in its own favour, nowhere as neat as Bostrom's theory.  By suggesting that the answer must rely on a perspective outside of our own dimension where the laws of origin are self-evident, we're already changing Bostrom's theory to something impossible to confirm through objective observation.  Although, if it were all done with computer code, thought itself would be no different than matter at its most basic component.  A supernova and a dream would essentially be composed of the exact same thing.
A final note.  If the speculation of theories like Bostrom's appear too strange to be true, it should be noted that superstring equations have been analysed and contain strings of what is called "doubly-even self-dual linear binary error-correcting block code," better known as an error correcting form of computer code invented by Claude Shannon in the 1940's.  At the 2011 Isaac Asimov memorial debate, Professor James Gates Jr. touched on the subject with colleagues and revealed the cryptic similarities the equations he found had with the same code running through the computers we have today leading him to, what he called, a strange place.